Fpga Implementation Of Low-light Video Clearness Algorithms

Under the influence of low lighting and poor equipments, the poor images from video camera can not satisfiy application requirements. The gray value of these images distributes a narrow range, so the contrast ratio of videos decreases and marginal information is fuzzy. Meanwhile, the video is dark since the average gray value is low. It is hard to find the target information in the video. Low illumination video clearness technology is very important in this environment. Because FPGA possesses rich logic resources and strong ability of real-time processing, it is very suitable to implement video enhancement algorithm. This manuscript developes a set of stable and reliable low illumination real-time video system based on FPGA.First, this manuscript simply introduces the research status of low illumination clearness algorithms and their hardware implementation, the shortcomings of existing technology is analyzed.Then, according to the requirements of real time video processing for stability, data signal synchronization technology is introduced. External DDR2 memory read/write rules are designed. In this paper, according to the characteristics of low illumination video that the average grey value is low and the gray level distribution is very concentrated, two low illumination video clearness algorithms are introduced. And the two existing algorithms are optimized and decomposed to come up with a hardware structure of real-time video clearness based on the FPGA.At last, in order to advoid the influence of the the highlight point on the video after low illumination enhancemet, this manuscript proposes an image fusion method based on inverse Gamma correction and inhibits the excessive enhancement phenomenon of highlight sections. The article discusses the real-time fusion of image enhancement algorithm and highlight restrain. The weight calculation procedure of fusion function is optimized.Real-time video experiments from industrial camera show that, the design of video acquiring system is stability and clear. The system can enhance the quality of video image significantly and also has certain inhibitory effect for highlight area.